"Build a man a fire, and he'll be warm for a day. Set a man on fire, and he'll be warm for the rest of his life." - Terry Pratchett

***

"The two major types of grading schemes are norm referenced (grading on a curve) and criterion based (meeting a set standard). Each type has both advantages and disadvantages. The advantages of grading on a curve are that it can discriminate among individuals, regardless of the overall ability level of the group, and will provide an even or predetermined grade distribution. It is the most popular educational measurement and, therefore, requires minimal faculty training. The disadantages of norm-referenced grading are that it is based on a relative standard that changes with the performance of each group and assesses students' status in relation to one another, not on their proficiency in subject matter. It leads to the self-fulfilling prophecy that some students will be high achievers, many will do moderately well, and a few will fail. Grading on a curve assumes a normal distribution of activity, but teaching is a purposeful activity. Because students are ranked against each other, curving grades encourages competition, instead oc ooperation. It increases students' anxiety about grades because grades cannot be determined until the end of the course. Additionally, it is demotivating for most students, because they achieve well on assignments but receive a poor grade depending on the skill of the competition. With norm-referenced grading, incompetent yet top-ranked students may get high grades. And, at the other extreme, a certain number of students must fail, even if they perform well.

Criterion-referenced grading has many advantages. For example, assessment is based on comparison against a standard, not on the performance of others. It is useful in selecting individuals who can perform at a given competence level because it indicates what students have learned, not their status in comparison to others. Criterion grading also can assess both teaching and learning if valid criterion levels and entering prerequisites exist. In addition, grade distribution is unaffected by an unusual number of high or low achievers in one class. It is therefore motivating for most students because all can earn top grades and success depends solely on actual achievement.

The disadvantages of criterion-referenced grading include that it is unfamiliar, thus requiring explanation, defense, and faculty training. It also is difficult to establish criterion levels. And because there is no automatic "weeding out" of a certain percentage of students, all students may fail or all may earn top grades."

--- Laurie Richlin, Blueprint for Learning